As online merchandise networks develop, more and more users are willing to make purchases online. During the purchasing process, the server on the network typically provides users with information of various kinds of products to satisfy the users' purchasing needs. Existing systems sometimes provide the users with product rankings by displaying products in a ranked list.
Given the massive amount of available data in large scale merchandise networks (in some instance data pertaining to billions of products), however, displaying a large number of products in a ranked list has become a problem. Users are often faced with an overwhelming number of products, many of which are of no interest to the user. The typical solutions used by systems today often neither satisfy the variety of user's needs nor enhance the purchasing experience for the user.